It's all about the answers!

Ask a question

Good working Practices around Test Artefacts Reviews and Approvals


Dan Evans (392810) | asked Oct 29 '12, 11:15 p.m.
 Hi,

We're looking at incorporating some good working practices around Test Artefact creation.

I want to be able to create a test plan 
          get the test plan reviewed.
          create test cases.
           review the test cases/approve
           once review then create new test scripts.

But it seems that the process to review has to be post-child artefact review in order to set it to read-only.

For example if I wanted the Test Cases reviewed and approved before the creation on a test script I would then have to unlock the 'approved' test case to add a test script to it.  

We are also going to have a rule which will link the Test Case 1 to 1 with test scripts so am I just being over zealous with the review and approve?  Or can I link the Test Case review to the Test Script review.

Any general advice surrounding how people manage the review and approval process and when they 'approve' parent artefacts would be gratefully received.
 
I don't want to reinvent the wheel so are there some standard policies that most people employ?


Many thanks,

Dan.



Accepted answer


permanent link
David Chadwick (1511) | answered Nov 01 '12, 11:34 a.m.
JAZZ DEVELOPER
I'm not sure that many of the internal IBM groups enforce approvals prior to execution.  The approval prior to execution does make audits easier if that is a requirement for your organization.  Overall getting peer reviewers for scripts and test cases is an overhead burden that must be included in your test group's ability to churn out more test cases and scripts as the review process adds significant ceremony and overhead time spent to the process.

Things to consider when determining your test process:
1. Is it a business requirement to have mandatory reviews and audit trails for all tests executed?
2. Is test coverage of all requirements more important than "deep" testing of complex features?
3. Is my test team experienced enough to maximize individual testing output or are peer reviews and test strategy sessions needed to get the desired testing results?
4. How much if any of my testing effort should be strategically directed towards areas of high defect density or new complexity to reduce overall risk?
5. Are my requirements specified with enough completeness and detail to test them as they are OR do I have to spend testing resources to get requirements elicited to a testable level?
6. What set of test environment factors must be covered and what strategy do I employ?

Once you have gauged what amount of testing resources you have to do your testing and how much testing has to be done, then you can choose your testing process in terms of its formality and the associated overhead cost.  Few organizations impose a strict process unless there is an industry mandate for it as there are never enough resources to test all that you need to test.

I realize that I failed to answer the question about testing approvals being batched, but it is really not a very important question as each approver gets notified of each approval request.  It goes in your work queue and you whittle them down as a pile on the days that you decide to take time out to do approvals.  Batching them up and holding a meeting is just not an efficient way to handle them.  It should be a background task that fits the approver's schedule and left with a date due and overdue reminder notices to get it cleaned up across a set of items.  At some point a team lead will have to send reminders to clean up approvals for a given milestone but otherwise it shouldn't be a team meeting activity requiring batching.  Obviously this is just my opinion and is colored by my experience with test teams having experienced and disciplined team members.

Dan Evans selected this answer as the correct answer

Comments
Dan Evans commented Nov 02 '12, 5:16 p.m.

Hi David,


Many thanks for your time for the in-depth answer.   This was better than the question that I asked.  Your insight has helped to answer a number of things.  
I'm putting in quite a bit of overhead in the formal reviews I will be expecting the teams to do. 

We do have requirements for audits and the cost of getting something wrong could be very expensive for us so I think its valid.  

As we are taking the approach of one RQM project for all delivery projects I figure Test Cases and Test Scripts will be reused more and so formal reviews will be more of an initial hit. 

Thanks very much on for your time.


Dan.




One other answer



permanent link
Paul Slauenwhite (8.4k12) | answered Oct 30 '12, 7:01 a.m.
FORUM MODERATOR / JAZZ DEVELOPER
Hi Dan,

Your QM process will either follow the default RQM process template and workflow or be customized for your Team/Department/Company.  Sounds like you need to define a custom workflow.

Paul

Comments
1
Dan Evans commented Oct 30 '12, 11:26 a.m.

  

Do people generally ensure the review/approval of a artefact before the start its child (Test Case/Test Script). Are there typical configurations (recipes) of workflows that most people follow.

I'm just wondering how people generally create their test cases and test scripts.  Is there a way you can bundle up the review of the script with the test case?

Many thanks,


Dan.

Your answer


Register or to post your answer.


Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.